Aviation power matrix consistent tracking control system and method thereof

Abstract

The invention discloses an aircraft power matrix consistent tracking control system and a method thereof, belongs to the field of aircraft engine control, and realizes collaborative thrust distribution and high-precision control of multiple aircraft engines in a complex environment by designing a dynamic consistent tracking strategy. The system comprises a piloting aero-engine unit, a plurality of following aero-engine units and a piloting-following communication network. Based on a distributed fixed time observation method, efficient observation of the state of the piloting aero-engine is realized; estimating and compensating unmodeled dynamics, uncertainty and external disturbance in real time based on a fixed time expansion state disturbance observation method; a dynamic inverse control law is designed through a non-smooth anti-interference backstepping control strategy, and it is ensured that the state of the following engine dynamically and consistently tracks the pilot engine. The system and the method have remarkable advantages in the aspects of improving the robustness, the anti-interference performance and the control precision of multi-aero-engine cooperative control, and can be widely applied to power systems of complex task aircrafts.

Description

Technical Field

[0001] The present invention belongs to the field of aircraft engine control and relates to a multi-aircraft engine coordinated control technology, and in particular to an aircraft power matrix consistent tracking control system and a method thereof, which are used to realize anti-interference consistent tracking control of multiple aircraft engines and ensure the dynamic and static performance and robustness of the control system. Background Art

[0002] Aircraft engines are the core power units of modern aircraft, and their performance is directly related to the operating efficiency, safety, and mission completion capabilities of aircraft. With the continuous development of modern aircraft technology, it is difficult for a single aircraft engine to meet the needs of aircraft, especially multi-task complex aircraft, in terms of thrust output and performance indicators. Multi-aircraft engine collaborative control has gradually become an important development trend. The aviation power matrix control system is a complex system formed by the cooperation of multiple aircraft engines. Each aircraft engine is an entity with autonomous decision-making capabilities. These entities communicate with each other and ultimately achieve the goal of consensus tracking control. Compared with traditional single-engine control, this distributed collaborative control architecture has higher reliability and fault tolerance. In the event of a single engine failure, it can still ensure the stable flight of the aircraft.

[0003] The consensus tracking control problem has been a research hotspot in multi-agent collaborative control in recent years. There have been many studies in the fields of drone formation control, traffic vehicle scheduling control, and unmanned ship cluster control. However, in the field of aviation motivation, there are currently few studies and related research is still in its infancy.

[0004] In fact, the pilot-follower type is one of the effective design solutions to the consistent tracking problem. In the pilot-follower aircraft engine system, the pilot aircraft engine can be a real aircraft engine or a command signal. Its purpose is to guide the state of the follower aircraft engine to converge to the predetermined target value, and multiple follower aircraft engines form an aviation power matrix control system and operate in coordination. Therefore, considering the pilot-follower aircraft engine control problem is more convenient for practical application. However, the coordinated control between multiple aircraft engines involves complex dynamic modeling, real-time interference compensation, distributed observation and communication, etc. The technical difficulty is relatively high and the practical application faces many challenges.

[0005] On the one hand, due to the complexity of the working environment of aircraft engines and the degradation of their own performance, the widely existing unmodeled dynamics, uncertainties and external disturbances in the system can all be attributed to the description of lumped disturbances. The existence of lumped disturbances will inevitably affect the control performance of aircraft engines. It is very challenging to achieve high-performance control of the aircraft power matrix anti-interference consistent tracking control system that meets stability, rapidity, accuracy, robustness and reliability. On the other hand, the aircraft engine itself has strong nonlinear dynamic characteristics, and its input-output relationship is difficult to describe by traditional linear control theory. In addition, due to the certain heterogeneity between the engines in the aircraft power matrix, how to achieve global consistent tracking through distributed control algorithms and how to effectively solve network problems such as communication delays and data packet loss are important technical difficulties.

[0006] Therefore, in response to the urgent technical problems faced by the current and future development of aviation engine control technology, how to design a high-performance, high-reliability aviation power matrix consistent tracking control method to make full use of engine model information, achieve accurate observation of the state of the pilot aircraft engine, real-time estimation and compensation of the lumped interference, and improve the dynamic and static performance, robustness and consistent tracking capability of the aviation power matrix control system is a technical problem that needs to be solved urgently in the current aviation power field. Summary of the invention

[0007] 1. Purpose of the invention In order to solve the above defects and deficiencies existing in the existing aircraft engine control technology and realize the tracking task of the pilot aircraft engine-follower aircraft engine system, the present invention aims to provide an aviation power matrix consistent tracking control system and method thereof. By adopting a pilot-follower design structure, a distributed observer based on fixed time is introduced to observe the state information of the pilot aircraft engine and an extended observer based on fixed time is used to observe the lumped interference. Further, based on the dynamic inversion method, the backstepping method and the non-smooth anti-interference technology, a non-smooth anti-interference backstepping controller is designed to realize the aviation power matrix consistent tracking control. This idea makes full use of the network communication between the control systems of various aircraft engines, generates the expected speed of each follower aircraft engine through a distributed observer, and designs a non-smooth anti-interference backstepping controller to make each follower aircraft engine track its own expected speed. This controller has the advantages of disturbance estimation and compensation, closed-loop stability, high control accuracy, and no reliance on accurate mathematical models. In theory, the closed-loop control system is consistently bounded and stable. Reasonable selection of observers and controller parameters can effectively improve the dynamic and static performance, anti-interference ability and consistent tracking ability of the aviation power matrix, and meet the high-precision, rapidity and robustness requirements of the aircraft for thrust distribution in complex mission scenarios.

[0008] (II) Technical solution In order to achieve the purpose of the invention and solve the technical problems, the present invention adopts the following technical solutions: The first invention objective of the present invention is to provide an aviation power matrix consistent tracking control system for distributed cooperative control of multiple aircraft engines, meet the high-precision control requirements of aircraft for thrust distribution in complex environments, and realize dynamic consistent tracking control between a pilot aircraft engine and a follower aircraft engine. The control system includes at least a pilot-follower communication network, a pilot aircraft engine unit, and a plurality of follower aircraft engine units. Specifically: The pilot-follower communication network performs information exchange between the pilot aircraft engine unit and each follower aircraft engine unit based on the network topology and communication protocol; The pilot aircraft engine unit includes at least one instruction generator and one backstepping controller, wherein: the instruction generator is used to generate dynamic instructions according to the overall mission requirements of the aircraft, and the backstepping controller is used to receive the dynamic instructions output by the instruction generator and generate a control signal in combination with the state vector of the pilot aircraft engine, so that the pilot aircraft engine outputs a target speed according to the dynamic instruction requirements; Each of the following aircraft engine units includes at least one distributed fixed-time observer, one fixed-time disturbance observer and one non-smooth anti-disturbance backstepping controller, wherein: The distributed fixed-time observer receives and observes the output state of the pilot aircraft engine based on the pilot-follower communication network, adopts a fixed-time state observation method to provide a speed observation value of the pilot aircraft engine, and uses it as the expected speed of the follower aircraft engine; The fixed-time disturbance observer observes and estimates the lumped disturbance existing in the following aircraft engine unit based on the fixed-time extended state observation method, and transmits the lumped disturbance estimation value to the non-smooth anti-disturbance backstepping controller; The non-smooth anti-interference backstepping controller is based on the output of the distributed fixed-time observer and the lumped interference estimate provided by the fixed-time interference observer. It generates a closed-loop control law and outputs a control signal through a backstepping method, a dynamic inverse method and a non-smooth control strategy to achieve state tracking of the leading aircraft engine by the following aircraft engine.

[0009] The second invention object of the present invention is to provide an aviation power matrix consistent tracking control method, based on the aviation power matrix consistent tracking control system provided by the first invention object above, the control method at least comprises the following steps when implemented: Using the input and output data of the pilot aircraft engine and the follower aircraft engine, the affine nonlinear dynamic models of the pilot aircraft engine and each follower aircraft engine are constructed through the system identification method, and then the control model of the aviation power matrix is ​​established. In addition, the cooperation relationship between multiple aircraft engines is considered, and the topological structure of the pilot-follower communication network is established to realize the state transmission and information interaction between the pilot aircraft engine and the follower aircraft engine. Step SS2: Trace instruction generation According to the overall mission requirements of the aircraft, the command generator in the pilot aircraft engine unit is based on the given reference command r d , through a transfer function G ( s ) to generate dynamic instructions that meet the index requirements x 0d , and find the dynamic instruction x 0d Differentiation with respect to time The pilot aircraft engine realizes its own state adjustment based on the backstepping controller and adjusts according to the dynamic instructions. x 0d and its time derivative Outputting the corresponding speed signal as the target state of the following aircraft engine; Step SS3: Distributed fixed-time observations The distributed fixed-time observer in each following aircraft engine unit uses the pilot-follower communication network to obtain the state information of the pilot aircraft engine and the adjacent following aircraft engines in real time, estimates and outputs the observed values ​​of the shaft speed and shaft acceleration of the pilot aircraft engine in real time based on the distributed fixed-time state observation method, and uses the obtained shaft speed observation value as the tracking instruction of the following aircraft engine, and provides the obtained shaft acceleration observation value to the subsequent dynamic control link as a reference signal; Step SS4: Fixed-time interference observation The fixed-time disturbance observer in each following aircraft engine unit estimates the lumped disturbance existing in the following aircraft engine unit in real time based on the fixed-time extended state disturbance observation method, and realizes efficient convergence of the error within a fixed time by introducing the synergistic effect of non-smooth terms and dynamic compensation terms. The lumped disturbance includes the influence of unmodeled dynamics, uncertainty and external disturbances, and the lumped disturbance estimation value is fed back to the non-smooth anti-disturbance backstepping controller; Step SS5: Non-smooth anti-interference backstepping control calculation The non-smooth anti-interference controller in each follower aircraft engine unit is designed based on backstepping design, dynamic inversion method and non-smooth anti-interference method to generate the closed-loop control law of the follower aircraft engine: As the first step of the backstepping design, according to the dynamic characteristics of the following aircraft engine unit and the shaft speed tracking error feedback, the expected value of the shaft acceleration of the following aircraft engine is designed as the intermediate target expected state of the recursive design, and it is smoothed by a non-smooth command filter to achieve dynamic consistent tracking of the output state of the leading aircraft engine. As the second step of the backstepping design, based on the dynamic inverse method, a non-smooth anti-interference backstepping control law is constructed by using the error between the current shaft acceleration of the following aircraft engine and the filtered expected shaft acceleration. The non-smooth anti-interference backstepping control law realizes dynamic tracking through linear and nonlinear error feedback, and introduces the lumped disturbance estimation value output by the fixed-time disturbance observer for compensation. The control signal is adjusted in real time according to the influence of unmodeled dynamics and external disturbances on system performance. At the same time, the boundedness and disturbance suppression capability of the control signal are ensured through the design of non-smooth functions. Step SS6: Closed-loop operation and control process judgment The control signal generated by the non-smooth anti-interference backstepping controller is applied to the following aircraft engine. At the same time, the system operation status is monitored in real time through the pilot-follower communication network. By confirming whether the state of each following aircraft engine has stably converged to the reference state of the pilot aircraft engine, it is judged whether the system meets the consistent tracking control target. If the consistent tracking control target is met, the control is terminated. If not, it returns to step SS2 and continues to execute the control cycle.

[0010] (III) Technical Effect Compared with the prior art, the aviation power matrix consistent tracking control system and method of the present invention have the following significant technical effects: (1) Not dependent on accurate models: The aviation power matrix of the present invention is composed of multiple follow-up aircraft engines. It only needs to use the input and output data of the aircraft engines and construct its nonlinear affine mathematical model with disturbance through the system identification method. The modeling error is described by lumped interference, and the lumped interference is observed and compensated in real time. Therefore, the accuracy of the model is not required to be high.

[0011] (2) Closed-loop stability: Compared with model-free control methods such as PID, this method can theoretically explain the bounded stability of the system. The present invention can more quickly and accurately describe and compensate for the internal and external factors of the aviation power matrix system through a fixed-time disturbance observer, thereby ensuring that the system is theoretically uniformly bounded and stable.

[0012] (3) Good control performance: Through the non-smooth anti-interference backstepping control method, while ensuring that the aviation power matrix control system is free from disturbances, it can also effectively achieve consistent tracking, taking into account dynamic and static performance and robustness. The method proposed in the present invention has good control performance.

[0013] (4) More importantly, the important innovation of this method lies in: proposing an aviation power matrix consistent tracking control method, considering the situation where the thrust of a single aircraft engine is limited, and providing the thrust required by the aircraft through the coordinated control of the aviation power matrix, and designing a leader-follower aircraft engine system to solve the aviation power matrix consistent tracking control problem. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a principle block diagram of an aviation power matrix consistent tracking control system provided by an embodiment of the present invention; Figure 2 is a flow chart of an aviation power matrix consistent tracking control method provided by an embodiment of the present invention; Figure 3 The figure shows the control effect diagram of the tracking step instruction of the aviation power matrix consistent tracking control method provided by the embodiment of the present invention; Figure 4 Shown is a control effect diagram of tracking sinusoidal instructions of the aviation power matrix consistent tracking control method provided by an embodiment of the present invention.

[0015] Explanation of the reference numerals: 0-leading aircraft engine unit, 01-command generator, 02-backstepping controller, 03-leading aircraft engine, 1-the 1st following aircraft engine unit, 11-the 1st distributed fixed-time observer, 12-the 1st fixed-time disturbance observer, 13-the 1st non-smooth anti-interference backstepping controller, 14-the 1st following aircraft engine, N-the Nth following aircraft engine unit, N1-the Nth distributed fixed-time observer, N2-the Nth fixed-time disturbance observer, N3-the Nth non-smooth anti-interference backstepping controller, N4-the Nth following aircraft engine. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. The described embodiments are part of the embodiments of the present invention, not all of the embodiments, and the described embodiments are exemplary and are intended to be used to explain the present invention, and cannot be understood as limiting the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0017] The aviation power matrix consistent tracking control system and method of the present invention are described in detail below in conjunction with specific embodiments. The present invention realizes dynamic consistent tracking control of multiple aircraft engines by designing a pilot-follower collaborative control strategy and a multi-module distributed control algorithm, aiming to solve the adaptability problem of traditional single thrust control systems in high-precision, multi-disturbance environments.

[0018] Example 1 As a specific example, Figure 1 As shown, the aviation power matrix consistent tracking control system provided by the embodiment of the present invention is used for the distributed cooperative control of multiple aircraft engines, meets the high-precision control requirements of the aircraft for thrust distribution in a complex environment, and realizes the dynamic consistent tracking control between the leading aircraft engine and the following aircraft engine. The system includes at least the following parts: The pilot aircraft engine unit 0 includes a command generator 01 and a backstepping controller 02, wherein the command generator 01 is used to generate dynamic commands according to the overall given index requirements. x 0d ; Backstepping controller 02, used to generate dynamic instructions according to the instruction generator 01 x 0d Produce control action, so that the pilot aircraft engine 03 follows the dynamic instruction x 0d Required output speed y 0 ; N following aircraft engine units with the same structure, taking the following aircraft engine unit 1 as an example, it includes a first distributed fixed time observer 11, a first fixed time disturbance observer 12, a first non-smooth anti-interference backstepping controller 13, and a first following aircraft engine 14, wherein the first distributed fixed time observer 11 is used to observe the output speed of the pilot aircraft engine y 0 The first fixed time interference observer 12 is used to observe the aggregate interference d 1 The first non-smooth anti-interference backstepping controller 13 realizes the output speed of the first aircraft engine 14 to the pilot aircraft engine 03 based on the output of the first distributed fixed time observer 11 and the first fixed time interference observer 12. y 0 Follow-up control; The pilot-follower communication network is used to complete mutual communication between the pilot aircraft engine unit 0, the first following aircraft engine unit 1, ..., and the Nth following aircraft engine unit N.

[0019] It should be noted that the present invention is mainly aimed at the speed control of multiple single-axis aircraft engines, including the pilot aircraft engine 03, the first following aircraft engine 14, ..., the Nth following aircraft engine N4. This part is an essential component of the present invention, but does not belong to the content of the present invention.

[0020] In the aviation power matrix consistent tracking control system of the present invention, the connection relationship between the various parts is as follows Figure 1 As shown. Among them, x 0 Describes the state vector of pilot engine unit 0 , u 0 is the output control quantity of the backstepping controller 02, y 0 is the output of the pilot aircraft engine unit 0 and , Indicates i Distributed fixed-time observer pair x 01 Observations of, i=1,2,…,N, x i Indicates i Following the state vector of the aircraft engine and , u i For the i The output control quantity of the non-smooth anti-interference backstepping controller, d i For the i Following the aggregate disturbance of the aircraft engine unit, Aggregate interference d i The estimated value of y i Indicates i Following the output of the aircraft engine unit and .

[0021] Example 2 Based on the aviation power matrix consistent tracking control system described in Example 1, this Example 2 describes in detail the aviation power matrix consistent tracking control method adopted by the system, such as Figure 2 As shown, the control method mainly includes the following steps when implemented: Step SS1: Establishment of aviation power matrix control model For a single-shaft aircraft engine system, the input and output data are used to obtain the affine nonlinear dynamic model of the aircraft engine through the system identification method for the needs of control design. NA control system following an aeroengine is established, and a network topology diagram of the leader-follower aeroengines is constructed, where the leader aeroengine is denoted as "0", and each follower aeroengine is denoted as " i "( i = 1, 2,.., N ). N The follower aeroengines together form an aero power matrix.

[0022] First, a dynamic model of the leader aeroengine is constructed to describe the relationship between the state vector of the leader aeroengine and the control input. Its state vector includes the shaft speed and shaft rotational acceleration, and the control input is the control signal output by the backstepping controller. The algorithm formula of the model is: (1) In the formula, x 0 represents the state vector of the leader aeroengine and , x 01 represents the shaft speed of the leader aeroengine, x 02 represents the shaft rotational acceleration of the leader aeroengine, are respectively x 01 , x 02 the differentials with respect to time and , are both non-linear functions with respect to the state vector x 0 , y 0 , u 0 represent the output value and control input quantity of the leader aeroengine respectively and .

[0023] The constructed dynamic model of the follower aeroengine is used to describe the relationship between the state vector, control input and lumped disturbance of the follower aeroengine. Its state vector includes the shaft speed and shaft rotational acceleration, the lumped disturbance includes unmodeled dynamics, uncertainties and external disturbances, and the control input is the output signal of the non-smooth disturbance rejection backstepping controller. The algorithm formula of the model is: (2) In the formula, , N is the number of follower aeroengines in the aero power matrix, x i , x i1 , x i2 , di Respectively represent i The state vector, shaft speed, shaft acceleration, and lumped disturbance of the aircraft engine are followed. , They are x i1 , x i2 Differentiation with respect to time and , All are about the state vector x i A nonlinear function of y i , u i Respectively represent i The output value of the aircraft engine, the control input and .

[0024] Step SS2: Trace instruction generation According to the overall mission requirements of the aircraft, the command generator in the pilot aircraft engine unit is based on the given reference command r d , through a transfer function G ( s ) to generate dynamic instructions that meet the index requirements x 0d , and find the dynamic instruction x 0d Differentiation with respect to time The pilot aircraft engine realizes its own state adjustment based on the backstepping controller and adjusts according to the dynamic instructions. x 0d and its time derivative The corresponding speed signal is output as the target state of the following aircraft engine.

[0025] It should be noted that the transfer function G ( s ) is designed to smooth the input reference command r d , thereby generating a command signal that meets the dynamic performance indicators x 0d . Through x 0d The time differential further provides a reference for the backstepping controller to dynamically adjust the speed, ensuring that the state of the pilot aircraft engine quickly approaches the target value.

[0026] Specifically, in aviation power matrix control, the dynamic instructions of the pilot aircraft engine x 0dIt is the key link for the whole system to achieve consistent tracking control. Directly use the reference instruction r d As a dynamic instruction, it may cause the system to respond too violently or oscillate, which is not conducive to the tracking performance of the aircraft engine. G ( s ) for reference instructions r d Shaping can effectively improve the dynamic characteristics of the system and ensure the smoothness and response speed of control.

[0027] Step SS3: Distributed fixed-time observations The distributed fixed-time observer in each follower aircraft engine unit uses the pilot-follower communication network to obtain the status information of the pilot aircraft engine and the adjacent follower aircraft engines in real time, and estimates and outputs the observed values ​​of the shaft speed and shaft acceleration of the pilot aircraft engine in real time based on the distributed fixed-time state observation method, and uses the obtained shaft speed observation value as the tracking instruction of the follower aircraft engine, and provides the obtained shaft acceleration observation value to the subsequent dynamic control link as a reference signal.

[0028] When the distributed fixed-time observers in each follower aircraft engine unit observe and dynamically update the shaft speed and shaft acceleration of the pilot aircraft engine in real time, the specific algorithm formula is: (3) (4) in, Respectively represent i , j The state observations of the following aircraft engine to the leading aircraft engine and , , Respectively i The following aircraft engine observes the shaft speed and shaft acceleration of the pilot aircraft engine. Respectively j The following aircraft engine observes the shaft speed and shaft acceleration of the pilot aircraft engine. They are Differentiation with respect to time, They are Differentiation with respect to time, i ≠ j and i , j = 1,2,…,N ; Respectively i , j The first follower aircraft engine to the leading aircraft enginem The observed values ​​of the state components and m =1,2; For the pilot aircraft engine The nonlinear dynamic characteristic function of α 1 , β 1 , γ 1 , α 2 , β 2 , γ 2 are appropriately selected parameters to ensure that the observation error converges to zero within a fixed time. μ is the adjustment parameter of the non-smooth term index and , ; a ij To pilot-follow the communication network i The first j The communication weight between the following aircraft engines, if the two communicate, then otherwise , Indicates i A follower aircraft engine cannot communicate externally with itself; b i For the i The communication weight between the follower aircraft engine and the pilot aircraft engine. If the two communicate, then ,otherwise .

[0029] Step SS4: Fixed-time interference observation The fixed-time disturbance observer in each follower aero-engine unit observes and estimates the lumped disturbance in the follower aero-engine unit in real time based on the fixed-time extended state disturbance observation method. By introducing the synergy of nonlinear high- and low-order feedback terms and dynamic compensation terms, the error is efficiently converged within a fixed time, and the lumped disturbance estimate is fed back to the non-smooth anti-disturbance backstepping controller. The lumped disturbance includes the influence of unmodeled dynamics, uncertainties, and external disturbances, such as parameter uncertainty caused by engine component wear, external disturbances such as high-frequency vibrations and airflow disturbances, and high-order dynamic characteristics ignored in the simplified modeling process.

[0030] When the fixed-time interference observer in each follower aircraft engine unit performs real-time estimation for the aggregate interference, the specific algorithm formula is designed as follows: (5) in, xi2 , d i Respectively i The following aircraft engine shaft acceleration, lumped disturbance, Corresponding to x i2 , d i The observed value of Indicates i The observation error of the shaft acceleration of the following aircraft engine; ρ 1 , ρ 2 and δ 1 , δ 2 are all exponential terms of non-smooth terms, and , ; Represents the observer gain, and all parameters are positive numbers; f i ( x i ), g i ( x i ) are all about x i Nonlinear function of .

[0031] Step SS5: Non-smooth anti-interference backstepping control calculation The non-smooth anti-interference controller in each follower aircraft engine unit is designed based on backstepping design, dynamic inversion method and non-smooth anti-interference method to generate the closed-loop control law of the follower aircraft engine: As the first step of the backstepping design, according to the dynamic characteristics of the following aircraft engine unit and the shaft speed tracking error feedback, the expected value of the shaft acceleration of the following aircraft engine is designed as the intermediate target expected state of the recursive design, and it is smoothed by a non-smooth command filter to achieve dynamic consistent tracking of the output state of the leading aircraft engine. As the second step of the backstepping design, based on the dynamic inverse method, a non-smooth anti-interference backstepping control law is constructed by utilizing the error between the current shaft acceleration of the following aircraft engine and the filtered expected shaft acceleration. The non-smooth anti-interference backstepping control law realizes dynamic tracking through linear and nonlinear error feedback, and introduces the lumped disturbance estimate output by the fixed-time disturbance observer for compensation. The control signal is adjusted in real time according to the impact of unmodeled dynamics and external disturbances on the system performance. At the same time, the boundedness and disturbance suppression capability of the control signal are ensured through the design of non-smooth functions.

[0032] Preferably, the non-smooth anti-interference backstepping control calculation is divided into two steps, respectively determining the desired state and actual control input of the following aircraft engine to cope with the unmodeled dynamics, uncertainties and external disturbances in the system, specifically: The first step is to design the expected state. This is done by designing the expected value of the shaft acceleration of the aircraft engine and using error feedback to achieve dynamic adjustment. The relevant algorithm formula is: (6) in, x i1d , x i2d Respectively i The following aircraft engine shaft speed x i1 , shaft acceleration x i2 The expected value of for x i1d Differentiation of time; K i1 , K i2 , K i3 are gain parameters used to adjust the error convergence speed and accuracy and are all positive numbers; u 1 , u 2 is the exponential term of the non-smooth term and , ; e i1 For the i The following aircraft engine shaft speed error and ; In order to solve the problem of "differential explosion" in actual systems, a non-smooth command filter is designed to generate smooth control commands to avoid direct use x i2d Possible instability: (7) in, are filter gain parameters and are all positive numbers. x i2c is the filter output value, i.e. the smoothed expected shaft acceleration, τ i2 is the time constant used to adjust the filter response speed; In the second step, based on the dynamic inversion method, the control law is designed by combining the outputs of the distributed fixed-time observer, the fixed-time disturbance observer and the non-smooth anti-interference backstepping controller to achieve consistent tracking control of the aviation power matrix. The relevant algorithm formula is: (8) in, For the i The following aero engine shaft acceleration x i2 The tracking error, K i4 , K i5 , K i6 are gain parameters used to ensure rapid error convergence and are all positive numbers. g i ( x i ) is about x i A nonlinear function of for x i2c Differentiation with respect to time, .

[0033] It should be noted that in the calculation of non-smooth anti-interference backstepping control, the dynamic inverse method is used to compensate for the nonlinear characteristics of the controlled object to ensure the dynamic consistent tracking of the aircraft engine; non-smooth anti-interference control achieves real-time suppression of unmodeled dynamics and external disturbances by introducing interference observations and nonlinear feedback mechanisms.

[0034] Step SS6: Closed-loop operation and control process judgment The control signal generated by the non-smooth anti-interference backstepping controller is applied to the following aircraft engine. At the same time, the system operation status is monitored in real time through the pilot-follower communication network. By confirming whether the state of each following aircraft engine has stably converged to the reference state of the pilot aircraft engine, it is judged whether the system meets the consistent tracking control target. If the consistent tracking control target is met, the control is terminated. If not, it returns to step SS2 and continues to execute the control cycle.

[0035] Figure 3The figure shows the control effect diagram of the tracking step instruction of the aviation power matrix consistent tracking control method of the present invention. The curves in the figure respectively represent the speed response of the pilot aircraft engine and multiple follower aircraft engines within the time range. It can be seen from the figure that at t=1s, the system gives a step speed instruction of 500 rpm. Under the action of the controller, the pilot aircraft engine can quickly track the given instruction and reach a steady state in about 1.5s. At the same time, although the initial speeds of the five follower aircraft engines are different, they can achieve state consistency within 0.5s under the synergistic effect of the distributed fixed time observer and the non-smooth anti-interference backstepping controller, and finally accurately track the speed instruction of the pilot aircraft engine. The entire control process shows good dynamic response characteristics, small overshoot, and steady-state error close to zero, which fully verifies the effectiveness of the aviation power matrix consistent tracking control method proposed by the present invention.

[0036] Figure 4 The figure shows the control effect diagram of the tracking sinusoidal command of the aviation power matrix consistent tracking control method of the present invention. The curves in the figure respectively represent the dynamic response of the pilot aircraft engine and multiple follower aircraft engines to the sinusoidal command. It can be seen from the figure that when the pilot aircraft engine tracks the sinusoidal command with an amplitude of 100rpm and a period of 2s, the five follower aircraft engines can achieve state consistency within 0.8s despite the large difference in initial speed, and accurately track the periodic motion trajectory of the pilot aircraft engine without obvious overshoot, which verifies the efficiency and robustness of the method of the present invention in dynamically tracking complex commands.

[0037] The above embodiments 1 and 2 have described in detail the aviation power matrix consistent tracking control system and method proposed in the present invention. Compared with the prior art, the aviation power matrix consistent tracking control system and method proposed in the present invention have significant technical effects such as independence from accurate models, closed-loop stability, good control performance, etc., and also have the following innovations: ① Aiming at the needs of collaborative work of multiple aircraft engines, an aviation power matrix consistent tracking control scheme is designed, and the aviation power matrix consistent tracking control problem is solved by establishing a pilot-follower aircraft engine system, so that multiple small-thrust aircraft engines can be combined into a large-thrust aviation power matrix; ② An aviation power matrix consistent tracking control method is proposed, and a distributed collaborative control algorithm is designed by utilizing network communication between aircraft engines. At the same time, the system lumped interference is estimated and compensated in real time, taking into account the dynamic and static characteristics of the system.

[0038] Through the above embodiments, the purpose of the present invention is fully and effectively achieved. Those skilled in the art can understand that the present invention includes but is not limited to the contents described in the drawings and the above specific embodiments. Although the present invention has been described with respect to the most practical and preferred embodiments currently considered, it should be understood that the present invention is not limited to the disclosed embodiments, and any modification that does not deviate from the functional and structural principles of the present invention will be included in the scope of the claims.

Claims

1. An aviation power matrix consistent tracking control system, comprising at least a pilot-follower communication network, a pilot aircraft engine unit and a plurality of follower aircraft engine units, characterized in that: The pilot-follower communication network performs information exchange between the pilot aircraft engine unit and each follower aircraft engine unit based on the network topology and communication protocol; The pilot aircraft engine unit includes at least one instruction generator and one backstepping controller, wherein: the instruction generator is used to generate dynamic instructions according to the overall mission requirements of the aircraft, and the backstepping controller is used to receive the dynamic instructions output by the instruction generator and generate a control signal in combination with the state vector of the pilot aircraft engine, so that the pilot aircraft engine outputs a target speed according to the dynamic instruction requirements; Each of the following aircraft engine units includes at least one distributed fixed-time observer, one fixed-time disturbance observer and one non-smooth anti-disturbance backstepping controller, wherein: The distributed fixed-time observer receives and observes the output state of the pilot aircraft engine based on the pilot-follower communication network, adopts a fixed-time state observation method to provide a speed observation value of the pilot aircraft engine, and uses it as the expected speed of the follower aircraft engine; The fixed-time disturbance observer observes and estimates the lumped disturbance existing in the following aircraft engine unit based on the fixed-time extended state observation method, and transmits the lumped disturbance estimation value to the non-smooth anti-disturbance backstepping controller; The non-smooth anti-interference backstepping controller is based on the output of the distributed fixed-time observer and the lumped interference estimate provided by the fixed-time interference observer. It generates a closed-loop control law and outputs a control signal through a backstepping method, a dynamic inverse method and a non-smooth control strategy to achieve state tracking of the leading aircraft engine by the following aircraft engine.

2. The aviation power matrix consistent tracking control system according to claim 1, characterized in that: In the aviation power matrix consistent tracking control system, the affine nonlinear dynamic models of the leading aircraft engine and each following aircraft engine are respectively constructed by the system identification method, and then the control model of the aviation power matrix is ​​established, wherein: The constructed dynamic model of the pilot aircraft engine is used to describe the relationship between the state vector and the control input of the pilot aircraft engine. The state vector includes the shaft speed and shaft acceleration. The control input is the control signal output by the backstepping controller. The algorithm formula of the model is: In the formula, x 0 represents the state vector of the pilot aircraft engine and , x 01 Indicates the shaft speed of the pilot aircraft engine, x 02 represents the shaft acceleration of the pilot aircraft engine, They are x 01 , x 02 Differentiation with respect to time and , f 0( x 0), g 0( x 0) are all about the state vector x 0 nonlinear function, y 0. u 0 represents the output value of the pilot aircraft engine, the control input and ; The constructed dynamic model of the following aircraft engine is used to describe the relationship between the state vector, control input and lumped disturbance of the following aircraft engine. Its state vector includes shaft speed and shaft acceleration. The lumped disturbance includes unmodeled dynamics, uncertainty and external disturbance. The control input is the output signal of the non-smooth anti-interference backstepping controller. The algorithm formula of the model is: In the formula, , N is the number of following aircraft engines in the aviation power matrix, x i , x i1 , x i2 , d i Respectively represent i The state vector, shaft speed, shaft acceleration, and lumped disturbance of the aircraft engine are followed. , They are x i1 , x i2 Differentiation with respect to time and , f i ( x i ), g i ( x i ) are all about the state vector x i A nonlinear function of y i , u i Respectively represent i The output value of the aircraft engine is followed by the control input value and .

3. The aviation power matrix consistent tracking control system according to claim 2, characterized in that: In each of the following aircraft engine units, when the distributed fixed-time observer observes and dynamically updates the shaft speed and shaft acceleration of the pilot aircraft engine in real time, the algorithm formula is: in, Respectively represent i , j The state observations of the following aircraft engine to the leading aircraft engine and , , Respectively i The following aircraft engine observes the shaft speed and shaft acceleration of the pilot aircraft engine. Respectively j The following aircraft engine observes the shaft speed and shaft acceleration of the pilot aircraft engine. They are Differentiation with respect to time, They are Differentiation with respect to time, i ≠ j and i , j = 1,2,…,N ; Respectively i , j The first follower aircraft engine to the leading aircraft engine m The observed values ​​of the state components and m =1,2; For the pilot aircraft engine The nonlinear dynamic characteristic function of α 1. β 1. γ 1. α 2. β 2. γ 2 are appropriately selected parameters to ensure that the observation error converges to zero within a fixed time. μ is the adjustment parameter of the non-smooth term index and , ; a ij To pilot-follow the communication network i The first j The communication weight between the following aircraft engines, if the two communicate, then ,otherwise , Indicates i A follower aircraft engine cannot communicate externally with itself; b i For the i The communication weight between the follower aircraft engine and the pilot aircraft engine. If the two communicate, then ,otherwise .

4. The aviation power matrix consistent tracking control system according to claim 3 is characterized in that: In each of the following aircraft engine units, when the fixed-time interference observer performs real-time estimation for the aggregate interference, the algorithm formula is specifically designed as follows: in, x i2 , d i Respectively i The following aircraft engine shaft acceleration, lumped disturbance, Corresponding to x i2 , d i The observed value of Indicates i The observation error of the shaft acceleration of the following aircraft engine; ρ 1. ρ 2 and δ 1. δ 2 are all exponential terms of non-smooth terms, and , ; Represents the observer gain, and all parameters are positive numbers; f i ( x i ), g i ( x i ) are all about x i Nonlinear function of .

5. The aviation power matrix consistent tracking control system according to claim 4, characterized in that: In each of the following aircraft engine units, the non-smooth anti-interference controller is divided into two steps when performing control calculations, respectively determining the desired state and actual control input of the following aircraft engine to cope with the unmodeled dynamics, uncertainties and external disturbances in the system, specifically: The first step is to design the expected state. This is done by designing the expected value of the shaft acceleration of the aircraft engine and using error feedback to achieve dynamic adjustment. The relevant algorithm formula is: in, x i1d , x i2d Respectively i The following aircraft engine shaft speed x i1 , shaft acceleration x i2 The expected value of for x i1d Differentiation of time; K i1 , K i2 , K i3 are gain parameters used to adjust the error convergence speed and accuracy and are all positive numbers; u 1. u 2 is the exponential term of the non-smooth term and , ; e i1 For the i The following aircraft engine shaft speed error and ; In order to solve the problem of "differential explosion" in actual systems, a non-smooth command filter is designed to generate smooth control commands to avoid direct use x i2d Possible instability: in, are filter gain parameters and are all positive numbers. x i2c is the filter output value, i.e. the smoothed expected shaft acceleration, τ i2 is the time constant used to adjust the filter response speed; In the second step, based on the dynamic inversion method, the control law is designed by combining the outputs of the distributed fixed-time observer, the fixed-time disturbance observer and the non-smooth anti-interference backstepping controller to achieve consistent tracking control of the aviation power matrix. The relevant algorithm formula is: in, For the i The following aero engine shaft acceleration x i2 The tracking error, K i4 , K i5 , K i6 are gain parameters used to ensure rapid error convergence and are all positive numbers. g i ( x i ) is about x i A nonlinear function of for x i2c Differentiation with respect to time, .

6. An aviation power matrix consistent tracking control method, based on the aviation power matrix consistent tracking control system according to any one of claims 1 to 5, characterized in that: The control method comprises at least the following steps when implemented: SS1. Establishment of aviation power matrix control model: Using the input and output data of the pilot aircraft engine and the follower aircraft engine, the affine nonlinear dynamic models of the pilot aircraft engine and each follower aircraft engine are respectively constructed through the system identification method, and then the control model of the aviation power matrix is ​​established. In addition, the cooperation relationship between multiple aircraft engines is considered, and the topological structure of the pilot-follower communication network is established to realize the state transmission and information interaction between the pilot aircraft engine and the follower aircraft engine; SS2. Tracking command generation: According to the overall mission requirements of the aircraft, the command generator in the pilot aircraft engine unit generates a given reference command based on the known reference command. r d , through a transfer function G ( s ) to generate dynamic instructions that meet the index requirements x 0d , and find the dynamic instruction x 0d Differentiation with respect to time The pilot aircraft engine realizes its own state adjustment based on the backstepping controller and adjusts according to the dynamic instructions. x 0d and its time derivative Output corresponding speed signal; SS3. Distributed fixed-time observation: The distributed fixed-time observer in each follower aircraft engine unit uses the pilot-follower communication network to obtain the state information of the pilot aircraft engine and the adjacent follower aircraft engines in real time, and estimates and outputs the observed values ​​of the shaft speed and shaft acceleration of the pilot aircraft engine in real time based on the distributed fixed-time state observation method, and uses the obtained shaft speed observation value as the tracking instruction of the follower aircraft engine, and provides the obtained shaft acceleration observation value to the subsequent dynamic control link as a reference signal; SS4. Fixed-time disturbance observation: The fixed-time disturbance observer in each follower aircraft engine unit estimates the lumped disturbance in the follower aircraft engine unit in real time based on the fixed-time extended state disturbance observation method, and achieves efficient convergence of the error within a fixed time by introducing the synergistic effect of non-smooth terms and dynamic compensation terms. The lumped disturbance includes the influence of unmodeled dynamics, uncertainty and external disturbances, and the lumped disturbance estimation value is fed back to the non-smooth anti-disturbance backstepping controller; SS5. Non-smooth anti-interference backstepping control calculation: The non-smooth anti-interference controller in each follower aircraft engine unit is designed based on backstepping design, dynamic inversion method and non-smooth anti-interference method to generate the closed-loop control law of the follower aircraft engine: As the first step of the backstepping design, according to the dynamic characteristics of the following aircraft engine unit and the shaft speed tracking error feedback, the expected value of the shaft acceleration of the following aircraft engine is designed as the intermediate target expected state of the recursive design, and it is smoothed by a non-smooth command filter to achieve dynamic consistent tracking of the output state of the leading aircraft engine. As the second step of the backstepping design, based on the dynamic inverse method, a non-smooth anti-interference backstepping control law is constructed by using the error between the current shaft acceleration of the following aircraft engine and the filtered expected shaft acceleration. The non-smooth anti-interference backstepping control law realizes dynamic tracking through linear and nonlinear error feedback, and introduces the lumped disturbance estimation value output by the fixed-time disturbance observer for compensation. The control signal is adjusted in real time according to the influence of unmodeled dynamics and external disturbances on system performance. At the same time, the boundedness and disturbance suppression capability of the control signal are ensured through the design of non-smooth functions. SS6. Closed-loop operation and control process judgment: The control signal generated by the non-smooth anti-interference backstepping controller is applied to the following aircraft engine. At the same time, the system operation status is monitored in real time through the pilot-follower communication network. By confirming whether the status of each following aircraft engine has stably converged to the reference state of the pilot aircraft engine, it is judged whether the system meets the consistent tracking control target. If the consistent tracking control target is met, the control is terminated. If not, it returns to step SS2 and continues to execute the control loop.

7. The aviation power matrix consistent tracking control method according to claim 6 is characterized in that: In step SS1, the dynamic model of the pilot aircraft engine is constructed to describe the relationship between the state vector and the control input of the pilot aircraft engine. The state vector includes the shaft speed and the shaft acceleration. The control input is the control signal output by the backstepping controller. The algorithm formula of the model is: In the formula, x 0 represents the state vector of the pilot aircraft engine and , x 01 Indicates the shaft speed of the pilot aircraft engine, x 02 represents the shaft acceleration of the pilot aircraft engine, They are x 01 , x 02 Differentiation with respect to time and , All are about the state vector x 0 nonlinear function, y 0. u 0 represents the output value of the pilot aircraft engine, the control input and ; The constructed dynamic model of the following aircraft engine is used to describe the relationship between the state vector, control input and lumped disturbance of the following aircraft engine. Its state vector includes shaft speed and shaft acceleration. The lumped disturbance includes unmodeled dynamics, uncertainty and external disturbance. The control input is the output signal of the non-smooth anti-interference backstepping controller. The algorithm formula of the model is: In the formula, , N is the number of following aircraft engines in the aviation power matrix, x i , x i1 , x i2 , d i Respectively represent i The state vector, shaft speed, shaft acceleration, and lumped disturbance of the aircraft engine are followed. , They are x i1 , x i2 Differentiation with respect to time and , All are about the state vector x i A nonlinear function of y i , u i Respectively represent i The output value of the aircraft engine is followed by the control input value and .

8. The aviation power matrix consistent tracking control method according to claim 7, characterized in that: In step SS3, when the distributed fixed-time observers in the following aircraft engine units observe and dynamically update the shaft speed and shaft acceleration of the pilot aircraft engine in real time, the specific algorithm formula is: in, Respectively represent i , j The state observations of the following aircraft engine to the leading aircraft engine and , , Respectively i The following aircraft engine observes the shaft speed and shaft acceleration of the pilot aircraft engine. Respectively j The following aircraft engine observes the shaft speed and shaft acceleration of the pilot aircraft engine. They are Differentiation with respect to time, They are Differentiation with respect to time, i ≠ j and i , j = 1,2,…,N ; Respectively i , j The first follower aircraft engine to the leading aircraft engine m The observed values ​​of the state components and m =1,2; For the pilot aircraft engine The nonlinear dynamic characteristic function of α 1. β 1. γ 1. α 2. β 2. γ 2 are appropriately selected parameters to ensure that the observation error converges to zero within a fixed time. μ is the adjustment parameter of the non-smooth term index and , ; a ij To pilot-follow the communication network i The first j The communication weight between the following aircraft engines, if the two communicate, then ,otherwise , Indicates i A follower aircraft engine cannot communicate externally with itself; b i For the i The communication weight between the follower aircraft engine and the pilot aircraft engine. If the two communicate, then ,otherwise .

9. The aviation power matrix consistent tracking control method according to claim 8, characterized in that: In step SS4, when the fixed-time interference observer in each follower aircraft engine unit performs real-time estimation for the aggregate interference, the algorithm formula is specifically designed as follows: in, x i2 , d i Respectively i The following aircraft engine shaft acceleration, lumped disturbance, Corresponding to x i2 , d i The observed value of Indicates i The observation error of the shaft acceleration of the following aircraft engine; ρ 1. ρ 2 and δ 1. δ 2 are all exponential terms of non-smooth terms, and , ; Represents the observer gain, and all parameters are positive numbers; f i ( x i ), g i ( x i ) are all about x i Nonlinear function of .

10. The aviation power matrix consistent tracking control method according to claim 9, characterized in that: In step SS5, the non-smooth anti-interference backstepping control calculation is divided into two steps, which respectively determine the desired state and actual control input of the following aircraft engine to deal with the unmodeled dynamics, uncertainties and external disturbances in the system, specifically: The first step is to design the expected state. This is done by designing the expected value of the shaft acceleration of the aircraft engine and using error feedback to achieve dynamic adjustment. The relevant algorithm formula is: in, x i1d , x i2d Respectively i The following aircraft engine shaft speed x i1 , shaft acceleration x i2 The expected value of for x i1d Differentiation of time; K i1 , K i2 , K i3 are gain parameters used to adjust the error convergence speed and accuracy and are all positive numbers; u 1. u 2 is the exponential term of the non-smooth term and ; e i1 For the i The following aircraft engine shaft speed error and ; In order to solve the problem of "differential explosion" in actual systems, a non-smooth command filter is designed to generate smooth control commands to avoid direct use x i2d Possible instability: in, are filter gain parameters and are all positive numbers. x i2c is the filter output value, i.e. the smoothed expected shaft acceleration, τ i2 is the time constant used to adjust the filter response speed; In the second step, based on the dynamic inversion method, the control law is designed by combining the outputs of the distributed fixed-time observer, the fixed-time disturbance observer and the non-smooth anti-interference backstepping controller to achieve consistent tracking control of the aviation power matrix. The relevant algorithm formula is: in, For the i The following aero engine shaft acceleration x i2 The tracking error, K i4 , K i5 , K i6 are gain parameters used to ensure rapid error convergence and are all positive numbers. g i ( x i ) is about x i A nonlinear function of for x i2c Differentiation with respect to time, .

Images